Inadequate follow-up of abnormal screening mammograms can result in delayed diagnoses of breast cancer which can lead to increased morbidity and mortality. Delays in diagnosis may also result in increased health care costs from the treatment of more advanced cancers and to substantial malpractice liability. The objective of this study is to determine the magnitude of the problem of inadequate follow-up and to determine why some women do not receive adequate follow-up. We will perform a retrospective cohort study of 1,000 women with abnormal screening mammograms cared for within the Henry Ford Medical Group, a large group practice that includes 130 staff physicians and residents delivering primary care. We will determine through billing data, chart reviews, and a telephone survey what proportion of women receive inadequate follow-up and determine what proportion of the cases of inadequate follow-up are due to failure to notify the patient of abnormal results. We will also identify problems in the process of patient notification. Using this cohort, we will conduct a population-based, case-control study design to identify, among women advised to have follow-up, risk factors for noncompliance based on data from the telephone survey. Patients without adequate follow-up and a sample of patients with adequate follow-up will be interviewed by interviewers blinded to their follow-up status and the subjects will be kept unaware of the purpose of the survey to avoid bias. Questions will elicit information about potential barriers to compliance as well as health beliefs and health behaviors. Finally we will survey all HFMG physicians delivering primary care by mailed questionnaires to determine what systems they use to ensure patient notification and follow-up. This study will provide information on the magnitude of inadequate follow-up. It will also serve to identity problems with patient notification and follow-up tracking systems as well as barriers to patient compliance which contribute to inadequate follow-up for some patients.